TFAM expression in brown adipocytes confers obesity resistance by secreting extracellular vesicles that promote self-activation

Figure 1. Effects of hTFAM overexpression on white and brown adipose tissue phenotypes (A) Verification of hTFAM protein expression in epididymal (e)- and inguinal (i)-WAT and BAT. (B) The h- and m- TFAM protein levels were evaluated using western blot analysis. Pictures represent western blots for h- and m- TFAM in BAT. (C) The ratio between hTFAM and mTFAM protein expression in Tg and TgTg BAT. (D) The h- and m- TFAM protein levels were normalized by the levels of HSP60. The quantitative results are expressed as the mean percentage of the level in the Tg ± standard error (SE) (n = 3/group). Total protein extracts from e− and i-WAT and BAT probed with a polyclonal antiserum against h- and m- TFAM. (E) Photographs show representative conditions of e − and i-WAT, liver, and BAT from NCD-fed WT, Tg, and TgTg mice. (F) Effect of hTFAM overexpression on e-adipocyte size. Pictures are representative of each sample stained with hematoxylin and eosin and examined by optical microscopy. Scale bar represents 100 μm. (G) Frequency distribution of the e-adipocyte area in NCD-fed WT, Tg, and TgTg mice. ∗∗p < 0.01, versus Tg mice. Two-tailed unpaired Student’s t test in (C) and (D).

Figure 2. Metabolic characterization in WT, Tg, and TgTg mice under a HFD (A) TgTg mice had extremely low e− and i-WAT accumulation on the HFD compared to WT mice. (B) Changes in body weight and food intake (n = 5–10/group). (C) Evaluation of insulin resistance (homeostasis model assessment of insulin resistance: HOMA-IR) and β cell function (insulinogenic index, homeostasis model assessment of beta-cell function: HOMA-β) (n = 5–10/group). (D) Glucose tolerance in 16 h-fasted mice after an intraperitoneal injection of glucose (5 mg/kg body weight); blood glucose and insulin levels measured at the indicated time points (n = 4–6/group). (E) Insulin sensitivity in 16 h-fasted mice after an intraperitoneal insulin tolerance test (0.5 units/kg body weight) (n = 4–6/group). (F) Microscopic e− and i-WAT and BAT sections stained with hematoxylin and eosin. Scale bars represent 100 μm. (G and H) VO2, VCO2, heat production, and RER were measured using an open-cage calorimetry system (n = 3/group). (I) Urinary levels of catecholamine, dopamine, noradrenalin, and adrenalin determined by high-performance liquid chromatography-mass spectrometry (HPLC-MS) (n = 4–6/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p…

Figure 3. Effects of hTFAM overexpression on BAT activation and WAT browning (A) Mitochondrial OXPHOS mRNA expression in brown adipose tissue. Total RNA was extracted from HFD-fed WT, Tg, and TgTg mice. All mRNA levels were measured by real-time RT-PCR. The mRNA levels were normalized to the levels of hypoxanthine-guanine phosphoribosyltransferase (Hprt), and the results are expressed as the mean percentage of the level in control WT mice ±SE (n = 3–4/group). (B) OXPHOS protein levels were evaluated using western blot analysis. Pictures represent western blots for OXPHOS in BAT and OXPHOS levels were normalized by the level of HSP90. (C)The quantitative results are expressed as the mean percentage of the level in the WT ± SE (n = 4–5/group). (D) Thermogenic and brown adipocyte differentiation/activation-related gene expression in BAT, as assessed by RT-PCR. (E and F) UCP-1 and PGC-1α protein levels in BAT were evaluated by western blot analysis (n = 4–9/group). The mRNA levels of beige adipocyte markers in i-WAT (G) and e-WAT (H) were assessed by RT-PCR (n = 7–10/group). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, versus WT mice. One-way ANOVA with Turkey’s multiple comparisons test in (A), (C),…

Figure 4. Effects of hTFAM-overexpressing brown adipocyte transplantation on metabolic characterization, BAT activity, and WAT browning in HFD-fed mice (A) Activated hTFAM-overexpressing brown adipocytes were injected subcutaneously around the BAT. Matrigel was confirmed around the BAT as a white substance. Pictures are representative of the Matrigel on the perimeter of the BAT at each time point (24 h and 14 d after injection); stained with hematoxylin and eosin. (# BAT, ∗Matrigel). Scale bars represent 500μm and 100 μm. (B) Immunostaining analysis of hTFAM-overexpressing brown adipocytes in Matrigel at each time point. Scale bar represents 50 μm. (C) Changes in body weight and food intake (n = 6–11/group). (D) Glucose tolerance was determined in 16 h-fasted mice after an intraperitoneal injection of glucose (5 mg/kg body weight) (n = 8–10/group). (E) Insulin sensitivity was determined in 16 h-fasted mice after an intraperitoneal insulin tolerance test (0.5 units/kg body weight) (n = 8–10/group). (F) Evaluation of insulin resistance (HOMA-IR) and β-cell function (HOMA-β) (n = 8–10/group). (G) Mitochondrial OXPHOS and (H) thermogenic and brown adipocyte differentiation/activation-related…

Figure 5. TgTg adipocytes secrete humoral factors and increase brown adipocyte activity (A) Spontaneous adipocyte differentiation in TgTg cells incubated with regular medium for 7 d. (B) Cross-section of the co-culture well. (C) Differentiated positive control TgTg adipocytes and aspects of WT adipocytes incubated in isolation or co-culture well. Number of mature cells was counted, and the results were expressed as the mean percentage of the number in WT adipocytes incubated in the isolation culture ± SE. ∗∗∗p< 0.001, versus isolation culture. (D) Brown adipocyte differentiation/activation-related gene expression assessed by RT-PCR. Levels of mRNA were normalized to the levels of Hprt, and the results are expressed as the mean percentage of the level in WT adipocytes incubated in the isolation culture ±SE (n = 3/group). ∗p < 0.05, ∗∗p< 0.01, versus isolation culture. (E) Differentiated WT cell numbers in the supernatants, transplanted from TgTg cell culture medium (n = 6/group). ##p < 0.01, ###p < 0.001, versus WT supernatants. One-way ANOVA with Turkey’s multiple comparisons test in (C). Two-tailed unpaired Student’s t test in (D) and (E). Bar graphs represent mean ± SE.

Figure 6. Enhanced EV release in TgTg brown adipocytes increases differentiation and activity (A) Differentiated positive control TgTg preadipocytes and aspects of WT preadipocytes incubated in 0.03 μm or 0.6 μm filtered co-culture well. Number of mature cells was counted, and the results were expressed as the mean percentage of the number in control WT adipocytes incubated in the isolation culture ±SE. ∗p < 0.05, versus 0.6 μm co-culture WT. (B) Thermogenic and brown adipocyte differentiation/activation-related gene expression in BAT, as assessed by RT-PCR. The mRNA levels were normalized to the levels of Hprt, and the results are expressed as the mean percentage of the level in control WT adipocytes incubated in the isolation culture ±SE (n = 10/group). ∗p < 0.05, ∗∗∗p < 0.001, versus control WT. (C) Molecular features of WT and TgTg brown adipocytes, assessed by electron microscopy. Scale bars represent 1 μm and 500 nm. High-power view of TgTg shows the MVB (boxed area). (D) Quantification of EVs derived from WT or TgTg brown adipocytes with the ExoCounter using anti-CD63 (left) or anti-CD9 (right) Ab-coated discs and beads conjugated with Ab against CD9 (n = 4/group). ###p < 0.001…